本研究模擬結果發現在學生學習動機方面，愈接近能力編班，學生可以在愈短的時間內使學習動機達到高點。在學生學習壓力方面，雖然能力編班下的學生提早承受學習壓力的高峰期，相對地，其課業成績也比其他編班方式的學生優異。因此，學生學習壓力與課業成績呈現正向關係。在教師教學壓力方面，不同編班策略下的教師所承受的教學壓力來源是不同的，在能力編班的班級中，教師最大的壓力是學生的課業成績；常態編班的班級中，教師主要的壓力來自課堂上學生的上課秩序和教學要求。 在無法改變教育政策的情況下，本研究嘗試探討在常態編班下影響學生課業成績的因素，結果以教師教學熱忱影響最大，降低教學壓力次之，而學生學習動機和學習成就感影響最小。 由於以往對於編班政策相關議題的研究方式，主要以一般社會科學的靜態研究為主。因此，本研究使用有別以往的研究方式，即系統動態學，運用Vensim套裝軟體進行模式建構與動態模擬。 From the results of simulation, this research found that, in the aspect of “learning motivation”, students placed in classes grouped by abilities reach the peak of learning motivation in shorter time spans, and the closer the levels of the students are in one class, the quicker they reach the peak of motivation. In the aspect of “learning stress”, students in classes grouped by abilities experience the peak of learning stress in a stage earlier than students in other classes; nonetheless, they still perform better in academic works. This shows that “learning stress” has a positive correlation to “academic performance”. In the aspect of “teaching stress”, teachers are subjected under different sources of teaching stress in different classes grouped by different strategies. In classes grouped by abilities, the highest level of teaching stress comes from the expectation projected on the teachers to maintain the students’ academic performance at a certain level. And in the randomly grouped classes, the highest level of teaching stress comes from the demand to maintain order and implement teaching in the classroom. Under the current education policies, this research makes an attempt to explore the factors that affect students’ academic performances. The results show that teachers’ enthusiasm has the most significant effect and lowering teaching stress ranks the second. Students’ “learning motivation” and “sense of achievement” appear to have the least effect among all other factors. Since the majority of the past studies were done through the static models from the aspect of social science, this research takes a different approach to survey the issues of class grouping from the aspect of system dynamics. Use of packaged software “Venism” is also incorporated into this study for model construction and dynamic simulation.
